This invention relates to a composition for gas treatment and processes and systems for making and using the composition. In particular, the invention relates to a high capacity regenerable sorbent for removal of mercury from flue gas and processes and systems for making and using the sorbent.
In December, 2000, the U.S. Environmental Protection Agency (EPA) announced its intention to regulate mercury and other air toxics emissions from coal- and oil-fired power plants with implementation as early as November, 2007 (Johnson, J., “Power Plants to Limit Mercury,” Chemical and Engineering News, 2001, p. 18, 79). The pending regulation has created an impetus in the utility industry to find cost-effective solutions to meet the impending mercury emission standards.
Mercury and its compounds are a group of chemicals identified in Title III of the 1990 Clean Air Act (CAA) Amendments as air toxic pollutants. Mercury is of significant environmental concern because of its toxicity, persistence in the environment, and bioaccumulation in the food chain. Mercury is one of the most volatile species of the 189 toxic compounds listed in the Clean Air Act Amendments and is, therefore, released readily into the environment from natural and anthropogenic sources. Because of its physical and chemical properties, mercury can also be transported regionally through various environmental cycles (Mercury Study Report to Congress, “Volume VIII: An Evaluation of Mercury Control Technologies and Costs,” U.S. Environmental Protection Agency, EPA-452/R-97-010, December, 1997). Atmospheric deposition of mercury is reported to be the primary cause of elevated mercury levels in fish found in water bodies remote from known sources of this heavy metal.
Domestic coal-fired power plants emit a total of about fifty metric tons of mercury into the atmosphere annually—approximately thirty-three percent of all mercury emissions from U.S. sources (Mercury Study Report to Congress, “Volume I: Executive Summary,” United States Environmental Protection Agency, EPA-452/R-97-010, December, 1997; Midwest Research Institute, “Locating and Estimating Air Emissions from Sources of Mercury and Mercury Compounds,” EPA-45/R-93-023, September, 1993). Specially designed emission-control systems may be required to capture these volatile compounds effectively. A coal-fired utility boiler emits several different mercury compounds, primarily elemental mercury, mercuric chloride (HgCl2), and mercuric oxide (HgO)—each in different proportions, depending on the characteristics of the fuel being burned and on the method of combustion. Quantifying the rate and composition of mercury emitted from stationary sources has been the subject of much recent work (e.g., Devito, M. S. et al., “Flue Gas Hg Measurements from Coal-Fired Boilers Equipped with Wet Scrubbers,” 92nd Annual Meeting Air & Waste Management Association, St. Louis, Mo., Jun. 21-24, 1999; Laudal, D. L. et al., “Bench and Pilot Scale Evaluation of Mercury Measurement Methods,” DOE/EPRI/EPA Joint Workshop on Mercury Measurement and Speciation Methods, Research Triangle Park, N.C., Jan. 29-30, 1997; Hargrove, O. W. et al., “Enhanced Control of Mercury by Wet FGD,” proceedings of First Joint Power and Fuel Systems Contractors Conference, Pittsburgh, Pa., Jul. 9-11, 1996; Agbede, R. O., A. J. Bochan, J. L. Clements, R. P. Khosah, T. J. McManus, “A Comparative Evaluation of EPA Method 29, the Ontario Hydro Method, and New Impinger Solution Methods for the Capture and Analysis of Mercury Species,” proceedings of the First Joint Power and Fuel Systems Contractors Conference, Pittsburgh, Pa., Jul. 9-11, 1996). Mercury is found predominantly in the vapor-phase in coal-fired boiler flue gas (Mercury Study Report to Congress, “Volume VIII: An Evaluation of Mercury Control Technologies and Costs,” United States Environmental Protection Agency, EPA-452/R-97-010, December, 1997). Mercury can also be bound to fly ash in the flue gas. Currently, there is no available control method that efficiently collects all mercury species present in the flue gas stream. Coal-fired combustion flue gas streams are of particular concern because of their composition that includes trace amounts of acid gases, including SO2, NO and NO2, and HCl. These acid gases have been shown to degrade the performance of activated carbon, the most widely available sorbent for mercury control, and other proposed sorbents, and so present a challenge that is addressed by the invention described herein.
Today, only municipal solid waste (MSW) incinerators and medical waste combustors are regulated with respect to mercury emissions, and, until the present, the best available control technology for these incinerators is the injection of activated carbon. Although fairly effective for MSW incinerators, activated carbon is a less appealing solution for coal-fired flue gas streams because of the dramatic difference in mercury concentrations. Regulations for mercury control from municipal and medical waste incinerators specify eighty percent control, or outlet emission levels of fifty micrograms per cubic meter (μg/m3). In coal-fired flue gas streams, typical uncontrolled mercury concentrations are on the order of ten μg/m3. For such low concentrations, projected injection rates for activated carbon to maintain ninety percent control of mercury emissions from coal-fired flue gas streams are high—up to 10,000 pounds or more of activated carbon to remove one pound of mercury, depending on the concentration and speciation of mercury in the flue-gas. The mercury-contaminated carbon becomes part of the ash collected by particulate-control devices and can convert the fly ash from an asset to a liability. This impact on the salability of collected fly ash can double the estimated cost of mercury control for a coal-fired power plant.
At present, the injection of activated carbon is broadly considered the best available control technology for reduction of mercury emissions from coal-fired power plants that do not have wet scrubbers (about seventy-five percent of all plants). Tests of carbon injection, both activated and chemically impregnated, have been reported by several research teams: Miller, S. J., et al., “Laboratory-Scale Investigation of Sorbents for Mercury Control,” paper number 94-RA114A.01, presented at the 87th Annual Air and Waste Management Meeting, Cincinnati, Ohio, Jun. 19-24, 1994; Sjostrom, S., J. et al., “Demonstration of Dry Carbon-Based Sorbent Injection for Mercury Control in Utility ESPs and Baghouses,” paper 97-WA72A.07, 90th Annual Meeting of the Air and Waste Management Association, Toronto, Ontario, Canada, Jun. 8-13, 1997; Bustard, C. J. et al., “Sorbent Injection for Flue-gas Mercury Control,” presented at the 87th Annual Air and Waste Management Meeting, Cincinnati, Ohio, Jun. 19-24, 1994; and Butz, J. R. et al., “Use of sorbents for Air Toxics Control in a Pilot-Scale COHPAC Baghouse,” 92nd Annual Meeting Air & Waste Management Association, St. Louis, Mo., Jun. 21-24, 1999. Activated carbon injection ratios for effective mercury control are widely variable and are explained by the dependence of the sorption process on flue gas temperature and composition, mercury speciation and also on fly ash chemistry.
The effectiveness of carbon injection for mercury control is limited by sorbent capacity and flue-gas interactions with the activated carbon. Flue gases contain several acid gases including sulfur dioxide (SO2) in the range of a few hundred to a few thousand parts per million (ppm); hydrogen chloride (hydrochloric acid, HCl) up to 100 ppm; and nitrogen oxides (e.g., NO2) in the range of 200 to 2,000 ppm. Studies reported by Miller, S. J et al., in “Mercury Sorbent Development for Coal-Fired Boilers,” presented at Conference on Air Quality: Mercury, Trace Elements, and Particulate Matter, McLean, Va., December 1998, at the University of North Dakota's Energy & Environmental Research Center (EERC) examined the effects of various acid gas constituents on the sorption capacity of carbon in a full-factorial test matrix. The EERC workers fed elemental mercury through carbon samples and systematically changed the gas composition. They noted a limited impact by SO2, but a dramatic drop in capacity when HCl or NO2 were present with SO2. Similar results were obtained in studies in a mercury test fixture by one of the applicants (Turchi, C., “Novel Process for Removal and Recovery of Vapor-Phase Mercury,” Final Report for Phase II, DOE Contract DE-AC22-95PC95257, Sep. 29, 2000). Thus, the instability of background art sorbents in an acidic flue gas environment adversely affects the utility of activated carbon sorbents and other sorbents having this limitation.
The cost to implement activated carbon mercury control systems has been estimated by the Department of Energy (DOE), EPA, and several researchers. Chang, R. et al., in “Mercury Emission Control Technologies,” Power Engineering, November, 1995, pp. 51-56, state that with operating and amortized capital costs, carbon injection will cost between $14,000 and $38,000 per pound of mercury removed, which equates to over $4 million per year for a 250-megawatt (MW) power plant.
EPA estimated similar costs for a 975-MW power plant (Mercury Study Report to Congress, “Volume VIII: An Evaluation of Mercury Control Technologies and Costs,” U.S. Environmental Protection Agency, EPA-452/R-97-010, December, 1997). In their model, four mercury control scenarios were considered to achieve ninety percent reduction in mercury emissions for a plant with an existing ESP. The scenarios were: (1) activated carbon injection; (2) spray cooling and activated carbon injection; (3) spray cooling, activated carbon injection with added fabric filter collection device; and (4) added activated carbon filter bed. The most economical control option employed spray cooling with carbon injection, resulting in a specific cost of $14,000 per pound mercury removed.
The development of more efficient sorbents that can function in the presence of acids would greatly reduce the cost of this mercury control approach by decreasing the amount of sorbent injected, downsizing sorbent injection equipment, and reducing costs for handling and disposing of spent sorbent.
The potential limitations of carbon-based sorbents cited above have led to research into other possible mercury sorbents. Research has demonstrated that noble-metal-impregnated alumina will remove elemental mercury and mercuric chloride from air streams. The sorbent can be thermally regenerated and the desorbed mercury captured in a condenser or oxidizing wet scrubber. Initial cost estimates looked attractive compared with the cost of disposable carbon sorbents (Turchi et al., “Removal of Mercury from Coal Combustion Flue-Gas Using Regenerable Sorbents,” 92nd Annual Meeting Air & Waste Management Association, St. Louis, Mo., Jun. 21-24, 1999). In 1998 and 1999, work at coal-combustion facilities in Pennsylvania and New Jersey demonstrated that the sorbent can function in flue-gas but at lower efficiency than was seen in the earlier laboratory tests. Subsequent lab work has suggested that acid-gas attack on the sorbent will reduce its effectiveness, as is the case for all background art sorbents. There is also some indication of flue-gas interactions similar to those witnessed by the EERC group. Research is continuing to determine whether the detrimental effects are temporary or permanent.
Fixed beds of zeolites and carbons have been proposed for a variety of mercury-control applications, but none has been developed specifically for control of mercury in coal flue-gas. Products in this class include Lurgi GmbH's (Frankfurt, Germany) Medisorbon and Calgon Carbon Corporation's (Pittsburgh, Pa.) HGR. Medisorbon is a sulfur-impregnated zeolite and costs ˜$17/lb. As with most sulfur-impregnated materials, Medisorbon loses sulfur when heated above 200° F., due to the vapor pressure of sulfur.
The background art is characterized by U.S. Pat. Nos. 3,194,629; 3,873,581; 4,069,140; 4,094,777; 4,101,631; 4,233,274; 4,474,896; 4,721,582; 4,814,152; 4,843,102; 4,877,515; 4,902,662; 4,911,825; 4,962,276; 4,985,389; 5,080,799; 5,120,515; 5,141,724; 5,173,286; 5,245,106, 5,248,488; 5,409,522; 5,695,726; and 5,846,434; the disclosures of which patents are incorporated by reference as if fully set forth herein.
No individual background art reference or combination of references teach the compositions, processes and systems disclosed herein. In fact, background art references teach away from the elegant solutions proposed herein.